September 2014 - ghc-tickets

[GHC] #8626: Bad magic. Expected: feedfacf, got: feedface
by GHC 07 Nov '14

07 Nov '14

#8626: Bad magic. Expected: feedfacf, got: feedface ----------------------------------+------------------------------------- Reporter: guest | Owner: Type: bug | Status: new Priority: high | Milestone: Component: GHCi | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: x86_64 (amd64) | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------------+------------------------------------- Loading package base-unicode-symbols-0.2.2.4 ... <interactive>: /Users/paulo/Library/Haskell/ghc-7.6.3/lib/base-unicode- symbols-0.2.2.4/lib/HSbase-unicode-symbols-0.2.2.4.o: Bad magic. Expected: feedfacf, got: feedface. ghc: panic! (the 'impossible' happened) (GHC version 7.6.3 for x86_64-apple-darwin): loadObj "/Users/paulo/Library/Haskell/ghc-7.6.3/lib/base-unicode- symbols-0.2.2.4/lib/HSbase-unicode-symbols-0.2.2.4.o": failed Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8626> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #9011: panic - interactiveUI:setBuffering2
by GHC 06 Nov '14

06 Nov '14

#9011: panic - interactiveUI:setBuffering2 ----------------------------------+------------------------------- Reporter: irneb | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: GHCi | Version: 7.6.3 Keywords: | Operating System: Linux Architecture: x86_64 (amd64) | Type of failure: GHCi crash Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------------+------------------------------- I've looked through some other tickets and tried installing libffi5, though in my distro (Kubuntu 13.10 x86-64) only libffi6 is available. Even after installing that I get the following: {{{ $ ghci GHCi, version 7.6.3: http://www.haskell.org/ghc/ :? for help ghc: panic! (the 'impossible' happened) (GHC version 7.6.3 for x86_64-unknown-linux): interactiveUI:setBuffering2 Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9011> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #9358: Improve flag description in the user guide
by GHC 06 Nov '14

06 Nov '14

#9358: Improve flag description in the user guide -------------------------------------+------------------------------------- Reporter: jstolarek | Owner: Type: bug | Status: new Priority: low | Milestone: 7.10.1 Component: Documentation | Version: 7.8.3 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Easy (less than 1 | Type of failure: hour) | Documentation bug Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- Sections 4.20 (Flag reference) and 4.10.2 (-f*: platform-independent flags) of the User Guide could use some updating: - `-floopification` is mentioned in 4.20 but not in 4.10.2 (which, sadly, is my fault), - Section 4.10.2 says: "These flags turn on and off individual optimisations. They are normally set via the -O options (...) The flags below are off by default, except where noted below." Does the last sentence mean they are on by default for `-O` or that they are enabled even with `-O0` (which is the default optimisation level)? - What is the default value of `-fmax-simplifier-iterations`? - 4.20 mentions `-fmax-simplifier-iterations`, `-fmax-relevant-bindings` and `-fmax-worker-args` but only the first one is described in 4.10.2 I suspect there are more inconsistencies. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9358> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 22

[GHC] #9211: Untouchable type variable (regression from 7.6)
by GHC 04 Nov '14

04 Nov '14

#9211: Untouchable type variable (regression from 7.6) ------------------------------------+------------------------------------- Reporter: simonpj | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- Oleg says: what used to type check in GHC 7.4.1 (and I think in 7.6.2, although I no longer have access to that version) fails in GHC 7.8.2. The following program type-checks with GHC 7.4.1 and GHC 7.8.2: {{{ {-# LANGUAGE RankNTypes #-} {-# LANGUAGE TypeFamilies #-} module T where foo :: (forall f g. (Functor f) => f a -> f b) -> [a] -> [b] -- foo :: (forall f g. (Functor f, g ~ f) => g a -> g b) -> [a] -> [b] foo tr x = tr x t = foo (fmap not) [True] }}} The following code (which differs only in the signature of foo) {{{ {-# LANGUAGE RankNTypes #-} {-# LANGUAGE TypeFamilies #-} module T where -- foo :: (forall f g. (Functor f) => f a -> f b) -> [a] -> [b] foo :: (forall f g. (Functor f, g ~ f) => g a -> g b) -> [a] -> [b] foo tr x = tr x t = foo (fmap not) [True] }}} type-checks with 7.4.1 but not with 7.8.2. The latter reports the error {{{ Couldn't match type `b' with `Bool' `b' is untouchable inside the constraints (Functor f, g ~ f) bound by a type expected by the context: (Functor f, g ~ f) => g Bool -> g b at /tmp/t.hs:12:5-25 `b' is a rigid type variable bound by the inferred type of t :: [b] at /tmp/t.hs:12:1 Expected type: Bool -> b Actual type: Bool -> Bool Relevant bindings include t :: [b] (bound at /tmp/t.hs:12:1) In the first argument of `fmap', namely `not' In the first argument of `foo', namely `(fmap not)' }}} Giving `t` the type signature `[Bool]` fixes the problem. Alas, I come across the similar untouchable error in situations where giving the type signature is quite difficult (in local bindings, with quite large types). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9211> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #9081: Template Haskell gets confused with scoped kind variables in a class declaration
by GHC 04 Nov '14

04 Nov '14

#9081: Template Haskell gets confused with scoped kind variables in a class declaration ------------------------------------+------------------------------------- Reporter: goldfire | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Template Haskell | Version: 7.8.2 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- When I say {{{ {-# LANGUAGE TemplateHaskell, DataKinds, PolyKinds, TypeFamilies #-} module Bug where import Data.Proxy $( [d| class kproxy ~ 'KProxy => C (kproxy :: KProxy a) where type TF (x :: a) :: Bool |]) }}} I get {{{ /Users/rae/temp/Bug.hs:7:4: Kind variable also used as type variable: ‘a_asJA’ In the declaration for class C_asJy /Users/rae/temp/Bug.hs:7:4: The exact Name ‘kproxy_asJB’ is not in scope Probable cause: you used a unique Template Haskell name (NameU), perhaps via newName, but did not bind it If that's it, then -ddump-splices might be useful }}} The code in the TH quote compiles fine on its own. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9081> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #8482: Permission denied when copying file to current directory with copyFile
by GHC 03 Nov '14

03 Nov '14

#8482: Permission denied when copying file to current directory with copyFile -------------------------------------+------------------------------------- Reporter: Henk-Jan | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: | Version: 7.6.3 libraries/directory | Operating System: Windows Keywords: copyFile | Type of failure: Incorrect result permission denied | at runtime Architecture: x86 | Test Case: Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | -------------------------------------+------------------------------------- While the MS-DOS command {{{ C:\tmp\b>copy c:\tmp\a\x }}} runs successfully, the Haskell equivalent {{{ copyFile "c:\\tmp\\a\\x" "" }}} results in {{{ copyFile.hs: .\.copyFile584.tmp: copyFile: does not exist (Het systeem kan het opgegeven pad niet vinden.) }}} The Dutch explanation says: "The system can not find the specified path" {{{ copyFile "c:\\tmp\\a\\x" "." }}} results in {{{ copyFile.hs: .\.copyFile3960.tmp: copyFile: permission denied (Toegang geweigerd.) }}} The Dutch explanation says: "Access denied" -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8482> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #9359: Deriving clause failure with polymorphic kinds
by GHC 03 Nov '14

03 Nov '14

#9359: Deriving clause failure with polymorphic kinds -------------------------------------+------------------------------------- Reporter: simonpj | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- Consider {{{ {-# Language GADTs, PolyKinds, TypeFamilies, DataKinds #-} module Fam where data Cmp a where Sup :: Cmp a V :: a -> Cmp a deriving (Show, Eq) data family CmpInterval (a :: Cmp k) (b :: Cmp k) :: * data instance CmpInterval (V c) Sup = Starting c deriving( Show ) }}} GHC 7.8.3 gives {{{ Fam.hs:12:13: Can't make a derived instance of ‘Show (CmpInterval ('V c) 'Sup)’: No family instance for ‘CmpInterval ('V c) 'Sup’ In the data instance declaration for ‘CmpInterval’ }}} which is obviously wrong. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9359> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #9633: PolyKinds in 7.8.2 vs 7.8.3
by GHC 03 Nov '14

03 Nov '14

#9633: PolyKinds in 7.8.2 vs 7.8.3 -------------------------------------+------------------------------------- Reporter: crockeea | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.3 Keywords: PolyKinds | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: GHC Blocked By: | rejects valid program Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- The following code '''compiles''' with GHC 7.8.2. This code has been distilled down from a larger example where I needed `-XPolyKinds` in Bar.hs but not in Foo.hs. In addition, the `modify` function is supposed to modify a mutable `Data.Vector`, hence the inner `go` function has the intended type of `Int -> m ()`, though strictly speaking it could return any value since everything is discarded by the `forM_`. {{{ -- {-# LANGUAGE PolyKinds #-} module Foo where import Control.Monad (forM_) import Control.Monad.Primitive import Bar -- Vector signatures unsafeRead :: (PrimMonad m) => v (PrimState m) a -> Int -> m a unsafeRead = error "type only" unsafeWrite :: (PrimMonad m) => v (PrimState m) a -> Int -> a -> m () unsafeWrite = error "type only" modify :: Int -> Bar v r modify p = Bar (p-1) $ \y -> do let go i = do a <- unsafeRead y i unsafeWrite y i a --return a forM_ [0..(p-1)] go }}} {{{ {-# LANGUAGE RankNTypes, PolyKinds #-} module Bar where import Control.Monad.Primitive data Bar v r = Bar Int (forall s . (PrimMonad s) => v (PrimState s) r -> s ()) }}} In 7.8.3, this above code results in the error (with `-fprint-explicit- kinds`) {{{ Foo.hs:19:23: Couldn't match type ‘a0’ with ‘r’ ‘a0’ is untouchable inside the constraints (PrimMonad m, (~) * (PrimState m) (PrimState s)) bound by the inferred type of go :: (PrimMonad m, (~) * (PrimState m) (PrimState s)) => Int -> m () at Foo.hs:(18,7)-(20,23) ‘r’ is a rigid type variable bound by the type signature for modify :: Int -> Bar * v r at Foo.hs:16:11 Expected type: v0 (PrimState m) a0 Actual type: v (PrimState s) r ... In the first argument of ‘unsafeRead’, namely ‘y’ In a stmt of a 'do' block: a <- unsafeRead y i Foo.hs:20:19: Couldn't match type ‘v1’ with ‘v’ ‘v1’ is untouchable inside the constraints (PrimMonad m, (~) * (PrimState m) (PrimState s)) bound by the inferred type of go :: (PrimMonad m, (~) * (PrimState m) (PrimState s)) => Int -> m () at Foo.hs:(18,7)-(20,23) ‘v’ is a rigid type variable bound by the type signature for modify :: Int -> Bar * v r at Foo.hs:16:11 Expected type: v1 (PrimState m) a0 Actual type: v (PrimState s) r ... In the first argument of ‘unsafeWrite’, namely ‘y’ ... Failed, modules loaded: Bar. }}} After much digging, I found that enabling `-XPolyKinds` in Foo.hs gives a more meaningful error: {{{ Foo.hs:19:23: Could not deduce ((~) (* -> k -> *) v v0) ... Expected type: v0 (PrimState m) a0 Actual type: v (PrimState s) r ... In the first argument of ‘unsafeRead’, namely ‘y’ In a stmt of a 'do' block: a <- unsafeRead y i Foo.hs:20:19: Could not deduce ((~) (* -> k -> *) v v1) ... Expected type: v1 (PrimState m) a0 Actual type: v (PrimState s) r ... In the first argument of ‘unsafeWrite’, namely ‘y’ In a stmt of a 'do' block: unsafeWrite y i a Failed, modules loaded: Bar. }}} Adding `PolyKinds` to Foo.hs ''and'' uncommenting `return a` results in the error: {{{ Foo.hs:17:12: Couldn't match kind ‘k’ with ‘k1’ because type variable ‘k1’ would escape its scope This (rigid, skolem) type variable is bound by the type signature for modify :: Int -> Bar k1 v r at Foo.hs:16:11-24 Expected type: Bar k1 v r Actual type: Bar k v0 r0 ... In the expression: Bar (p - 1) $ \ y -> do { let ...; forM_ [0 .. (p - 1)] go } ... Foo.hs:18:7: Kind incompatibility when matching types: v0 :: * -> k -> * v1 :: * -> * -> * ... When checking that ‘go’ has the inferred type ‘forall (k :: BOX) (m :: * -> *) b (v :: * -> * -> *) (v1 :: * -> * -> *). ((~) (* -> k -> *) v0 v, (~) k r0 b, (~) (* -> k -> *) v0 v1, PrimMonad m, (~) * (PrimState m) (PrimState s)) => Int -> m b’ Probable cause: the inferred type is ambiguous In the expression: do { let go i = ...; forM_ [0 .. (p - 1)] go } ... Failed, modules loaded: Bar. }}} These errors can be resolved by modifying the original code above in any of the following ways: 1. Remove `-XPolyKinds` from Bar.hs 2. Add an explicit kind signature to the `v :: * -> * -> *` parameter of type `Bar` 3. With `PolyKinds` in Bar but *not* Foo, uncommenting `return a` make GHC 7.8.3. compile What I'm trying to understand is 1. Why there is different behavior from 7.8.2 to 7.8.3. There doesn't seem to be anything addressing `PolyKinds` in the [http://www.haskell.org/ghc/docs/7.8.3/html/users_guide/release-7-8-3.html release notes vs 7.8.2] . 2. In the event the above code should not compile with 7.8.3, there error message could be much clearer. The first error message above didn't help me understand that `PolyKinds` was to blame, while the second error did a better job, and the third was very clearly a `PolyKinds` issue (i.e. `kind mismatch`) 3. For the third resolution option above, I can't see any reason that adding `return a` to the inner `go` function should make the code compile while leaving it out makes the code somehow ambiguous. This, if nothing else, seems like a bug. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9633> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #9433: Partially applied type family allowed but unusable
by GHC 03 Nov '14

03 Nov '14

#9433: Partially applied type family allowed but unusable -------------------------------------+------------------------------------- Reporter: hesselink | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler (Type | Version: 7.9 checker) | Operating System: Keywords: | Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: GHC Difficulty: Unknown | accepts invalid program Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- I was playing around with the following code, which tries to map a type family over a type (something I thought would not be allowed): {{{#!hs {-# LANGUAGE TypeFamilies , KindSignatures #-} type family Id x :: * type instance Id a = a type family Map (f :: * -> *) x :: * type instance Map f [a] = [f a] x :: Map Id [Bool] x = [] }}} Both GHC 7.8.3 and the current HEAD (7.9.20140811) accept this program! However, changing the definition of `x` to `[True]` gives an error: {{{ Couldn't match type ‘Id Bool’ with ‘Bool’ Expected type: Map Id [Bool] Actual type: [Bool] In the expression: [True] In an equation for ‘x’: x = [True] }}} If I define `y :: Id Bool` with value `True` (which works fine) and define `x = [y]`, I even get this error: {{{ Couldn't match type ‘Bool’ with ‘Id Bool’ Expected type: Id Bool Actual type: Id Bool In the expression: y In the expression: [y] }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9433> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #8960: SpecConstr usage explodes beyond 4GB with GHC 7.8.1 rc 2
by GHC 03 Nov '14

03 Nov '14

#8960: SpecConstr usage explodes beyond 4GB with GHC 7.8.1 rc 2 -----------------------------------+--------------------------------------- Reporter: MichalGajda | Owner: Type: bug | Status: new Priority: normal | Milestone: 7.8.1 Component: Compiler | Version: 7.8.1-rc2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time crash Unknown/Multiple | Test Case: cabal install hPDB Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | -----------------------------------+--------------------------------------- GHC 7.8 rc2 exhausts all memory on `Bio.PDB.EventParser.PDBEventParser` module in -O3 mode, but not in -O2 mode. The tail of ''-v3'' input seems to indicate that it dies during SpecConstr phase. The module successfully compiled with below 2GB RAM since GHC 6.12, so I wonder why the memory use exploded with the most recent version. *** SpecConstr: ==================== SpecConstr ==================== Result size of SpecConstr STEPS TO REPRODUCE: 1. `cabal install hPDB` RESULT EXPECTED: Successful compilation and installation CURRENT RESULT: Exhausts available memory and crashes when compiling `Bio.PDB.EventParser.PDBEventParser`. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8960> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9